ELSEVIER SHORT NOTE 0006-3207(94)00041-7 B&logical Conservation 71 (1995) 205 207 ~ 1995 Elsevier Science Limited Printed in Great Britain. All rights reserved 0006-3207/95/$9.50 THE OPTIMIZATION OF BIODIVERSITY CONSERVATION Lars Witting & Volker Loeschcke Department of Ecology and Genetics, University of Aarhus, Ny Munkegade, DK-8000 Aarhus C, Denmark (Received 14 June 1993; revised version received 17 May 1994: accepted 20 May 1994) Abstract Biodiversity conservat&n & confronted with two major problems." how to define and measure biodiversity, and how to optimize the in situ conservation of biodiversity. Here we outline a conceptual framework for biodiversity conservation that is directed towards these problems. The framework combines a phylogenetic evaluation with a multi-species risk analysis and defines the objective of conservation biology as the minimization of the future loss of biodiversity. Keywords: conservation, phylogenetic diversity, opti- mization, nature reserve, species interactions. similar approach was mentioned by Weitzman (1992), but he did not describe how to estimate the expected loss of diversity. In the first section of the paper we describe how the biodiversity that is contained within a set of species can be evaluated from the phylogeny of that species set. This idea was first addressed by Vane-Wright et aL (1991) and later elaborated by oth- ers (e.g. May, 1990; Erwin, 1991; Williams et al., 1991; Brooks et al., 1992; Crozier, 1992; Faith, 1992, 1994; Weitzman, 1992; Crozier & Kusmierski, 1994; Witting et al., 1994). We use an evaluation similar to that applied by Faith (1992). We then describe how to estimate the expected loss of phylogenetic diversity and, finally, we give a brief discussion. INTRODUCTION Recently several workers have developed methods to optimize reserve selection so that a maximal number of species can be contained within a reserve system (Kirk- patrick, 1983; Ackery & Vane-Wright 1984; Margules et al., 1988; Rebelo & Siegfried, 1990; Vane-Wright et al., 1991; Pressey et al., 1993). In this way, reserves can be established to contain as many species as possible. However, the proposed methods will provide optimal protection to overall biodiversity only when all species outside the reserves are equally vulnerable to threats, and all species within the reserves are equally protected against threats. In many situations these conditions will not hold as, for instance, among the avifauna of the Polylepis woodlands of the Andean highlands (Fjelds~t, 1993). There, an optimization of the biodiversity that is contained within reserves can jeopardize the overall conservation of biodiversity by using limited resources to collect common species into reserve systems while vulnerable species are left unprotected. To avoid this problem conservation efforts should be directed explicitly toward a minimization of the future loss of biodiversity from the overall area. In the present paper we describe how to estimate the expected loss of biodiversity. This is done by combin- ing a phylogenetic evaluation with the degree of vulner- ability to a species (its probability to become extinct). The framework is described in more detail by Witting et al. (1994) and by Witting et al. (in prep.). A rather 205 MEASURING BIODIVERSITY Vane-Wright et al. (1991) addressed the issue that the phylogeny of a set of species reflects the evolutionary diversity of that species set. Each species differs in the amount of evolutionary information that is unique to it compared with that which it shares with other species. In Fig. l(a) species A and B have four unique genes each. Species A and B also have eight unique genes in common. In general, the weight of a phylogenetic inter- node gives the uniqueness, or diversity, shared by all the species terminal to that internode. The phylogenetic diversity, D, of a species set is then given by the sum of internode weights D : ~ Di, (1) i=1 where D i is the weight of the ith internode of the phylo- genetic tree, and n is the number of internodes common to the phylogenetic tree. CONSERVING BIODIVERSITY To optimize the conservation of biodiversity we need to minimize the loss of phylogenetic diversity. This can be done by estimating the amount of phylogenetic diver- sity expected to be lost for each available conservation strategy. The strategy with the lowest loss will conserve most phylogenetic diversity. To estimate the expected